THE AUSTRALIA'S BOOK HERITAGE RESOURCES PROJECT€¦ · · 2015-05-12THE AUSTRALIA'S BOOK HERITAGE RESOURCES PROJECT IAN MORRISON ... but only South Australia used a standard MARC

THE AUSTRALIA'S BOOK HERITAGE RESOURCES PROJECT

IAN MORRISON

THE AUSTRALIA'S BOOK HERITAGE RESOURCES (ABHR) Project is building on the work carried out by the Australia and New Zealand Early Imprints Project (EIP). It aims to record on the Australian Bibliographic Network (ABN) all books printed before 1801 currently held in Australian libraries. This article outlines the progress of the ABHR Project and its implications for librarians and researchers.

The Early Imprints Project The origins and objectives of the Early Imprints Project were described in the early 1980s by Alan Brissenden and, in more detail, by Brian McMullin.1 In brief, the project set out to record all books printed before 1801 held in Australian and New Zealand libraries. It was a collaborative effort, administered on a state-by-state basis, co-ordinated through the major public and university libraries. In the late 1970s paid research assistants and volunteers began recordin~ the holdings of libraries and private collectors on specially prepared cards. These cards were gathered and stored at the various co-ordinating institutions.3

A significant br-product of this activity was the compilation of checklists of STC4 and Wing items, published between 1977 and 1983 in the BSANZ Bulletin.

The Australian records were stored in various ways in various locations: in Queensland, South Australia, Tasmania and Western Australia the records were stored in electronic format, but only South Australia used a standard MARC format. 6 In Tasmania, a printed catalogue was produced from the EIP database but never published. The bulk of the records that had been created on cards in New South Wales and the Australian Capital Territory were transferred to Monash University: these cards were sorted into English and non-English imprints sequences and interfiled by title, but no further processing was undertaken due to lack of funding. A few thousand cards were retained at the National Library, and a few thousand more at the University of Sydney. By the mid-1980s, when funding and formal collaboration ceased, recording was essentially complete in New Zealand and in all Australian states except New South Wales.

Ironically, at the beginning of EIP Sydney had been considered to be well advanced, largely due to the work of Rose Smith and John Fletcher. 7 The existence of this preliminary work appears to have engendered a sense of complacency, and the New South Wales arm ofEIP expended most of its energy recording smaller collections (in particular the Catholic libraries) whilst all but ignoring the State Library of New South Wales and leaving a significant

BSANZ Bulletin, v.17 no.3, 1993

114 fan Morrison

proportion of the vast holdings of the Fisher Library unrecorded. The ABHR Project will go a considerable way towards redressing this situation, but it is clear that some thousands of items in these two major institutions will remain to be recorded after the cessation of the AB HR Project.

Inception of the ABHR Project At a meeting sponsored by the Australian Academy for the Humanities in 1991 substantial revisions of the EIP project's methods and objectives were agreed to, along with a new name, which highlights the practical outcomes of the project: early printed books are a valuable intellectual resource in their own right, as well as being a significant part of our national heritage. An application was prepared for Australian Research Council funding under Mechanism Cs for $270,000 in 1992, with a further application for funding to complete the project in 1993 foreshadowed. The application was prepared by a consortium which consisted of the Australian Academy for the Humanities, the National Library of Australia, Monash University, the State Library of New South Wales, the University of Sydney, the Australian National University, and the Bibliographical Society of Australia and New Zealand; the project was to be coordinated by the Centre for Bibliographical and Textual Studies at Monash, where the great bulk ofEIP cards were housed. The ARC granted $200,000 for 1992 and $250,000 for 1993. Funding will cease at the end of 1993.

Four members of the consortium are active participants on a day-to-day basis in the ABHR Project. They are: Monash University, the University of Sydney, the State Library of New South Wales, and the National Library of Australia. The Centre for Bibliographical and Textual Studies at Monash University is the co-ordinating body. Each of the participating institutions has contributed resources to the project. ABN has waived its usual charges; Monash University has provided, from a range of sources, additional financial and logistical support; the University of Sydney has contributed additional staff time to the project; and the National Library and the State Library of New South Wales have provided logistical support.

The ABHR Project has two distinct but complementary aims. The first is to load the records gathered by EIP onto ABN. The second is to record on ABN the collections left unrecorded by EIP. It is hoped eventually to produce a CDROM. Discussions regarding this are, however, at a very preliminary stage.

The Australian Bibliographic Network In order to explain the operation of the ABHR Project it is necessary to explain the workings of ABN in some detail. ABN was established as a national shared cataloguing system by the National Library of Australia in the early 1980s, using software purchased from the Washington Library Network (WLN). The emphasis in the early stages of development was on current cataloguing. Rare books were not a priority, and the retrieval capacity of the WLN software was


The ABHR Project 115

severely limited: in particular, imprint details could not be searched.9 Even for general cataloguing purposes, it has been said, ABN 'is not "user-friendly". It is not even librarian-friendly, and requires a significant amount of training to be used effectively.' 10 In order to explain the operation of the ABHR Project it is necessary, therefore, to explain the workings of ABN in some detail.

The ABN database consists of three interlinked but distinct parts or 'files': the bibliographic file, the holdings file, and the vocabulary (or authority) file. The bibliographic file contains the actual catalogue records; the holdings file contains the holdings statements attached to the bibliographic records by the participating institutions; the vocabulary file contains the author and subject headings attached to the bibliographic records. The vocabulary file is intended to establish a controlled vocabulary for personal and corporate author headings, uniform titles, and topical, geographic and personal and corporate name subject headings.

Institutions which subscribe to ABN are allocated a unique identifying symbol, which can be attached to bibliographic records in the database. The first letter of this symbol identifies the state in which the institution is located, and the remaining letters are an acronym for the institution: for example, the State Library of Victoria is identified by the symbol 'VSL'. This institutional symbol appears automatically at the head of holdings statements entered under the institution's sign-on: all that is entered manually is the classification number or other location indicator used by the institution. Institution-specific data (for example, about bindings or imperfections in the copy to hand) can also be entered onto ABN under the institution's sign-on: this information, however, is accessible only to the institution that has entered it.

In creating a new bibliographic record the following steps are required. First, an 'interim' record must be created. Interims are intended as temporary markers, a brief description of the item being catalogued, and can be searched on the database immediately they are input. The WLN software requires the existence of an interim record before it will allow a full catalogue record to be entered. The aim is to prevent duplication of effort: a cataloguer finding an interim record on the database can see that someone else has commenced work cataloguing that item, and need not proceed further.

When the interim record has been entered, a full record can be created. This step involves the identification or creation of appropriate author and subject headings,ll and the strict application of MARC tags and codes. Unlike interims, full records do not appear on the database immediately they are input: they are processed in an overnight tape run and appear on the database the following day.

ABHR procedures Project staff based at Monash are entering data from EIP cards onto ABN and in addition have listed three collections not previously recorded by EIP. The AB HR Project has also entered into an agreement with Monash University Library to


116 fan Morrison

share the costs of cataloguing the Monash Rare Books backlog: this work has been contracted out to CAVAL. 12

At the State Library of New South Wales and the Fisher Library, University of Sydney, staff are cataloguing material left unrecorded by EIP.

The National Library has used the EIP data from its own collections to upgrade the records created by its recent retrospective conversion, and it is now creating original records for the material not recorded by EIP. The ABN Office is arranging for the conversion and loading of tapes of EIP records from South Australia, Queensland, Western Australia and Tasmania. The South Australian records were loaded in late 1992;13 at the time of writing Guly 1993) the date of conversion and loading of the remaining records is uncertain.

Records added to the ABN holdings file by the ABHR Project appear under the generic prefix DBM:J, followed by the symbol for the individual institution. At the completion of the project, the holdings recorded by ABHR will be transferred to the symbols of the individual institutions.

Technological developments in the late 1980s The objectives and priorities of the ABHR Project differ in small but significant ways from those of EIP. These differences have a great deal to do with technological developments over the last decade.

The eventual outcome of EIP was always envisaged as some form of machine readable catalogue. The cards prepared for data collection were set out in MARC format, and the records gathered in several states were keyed into various electronic formats. At the time of the inception of EIP, however, there was no national bibliographic database; by 1983, when EIP had done the bulk of its work, ABN was still in the early stages of its development; and even by 1986, when EIP activity had practically ceased, access to ABN was still limited.

Several significant developments over the last few years have greatly enhanced the usefulness of ABN as a mechanism for cataloguing early printed books. Firstly, there have been a number of large retrospective conversion projects, the last and most significant for AB HR being the National Library's in 1989 and 1990. Retrospective conversions are always a mixed blessing, as they inevitably result in poor-quality records - often with non-standard forms of headings - but they do add to the total stock of bibliographic information available on the database. Another development of this kind has been the loading of OeLC14

records for microfilm copies of Wing and early American material. A number of recent software developments have further enhanced the

capability of ABN. The development of Supersearch has made the data recorded on ABN much more accessible: of particular relevance to early printed books is the capacity to search on imprint details and on bibliographic references. Where the original WLN software will search only limited areas of a record, the ST AIRS15 software used by Supersearch allows any part of a record' to be



se~ched. It is now possible, for example, to search the database for a specific STC or Wing number .16

Another recent enhancement, which was quite crucial for the viability of the ABHR Project, is the 'clone' command for close-copy cataloguing. This command allows an existing record to be used as the basis for a new one: the only re-keying required is of the details that are actually different in the new item. In cataloguing different editions of the same work, or a series of very similar works (e.g. almanacs), 'cloning' can drastically reduce the keyboarding time for each item, as well as the opportunity for transcription errors. The OCLC microfilm records noted above have thus become a major source for the AB HR Project: in cataloguing the vast majority of Wing items, all that is required is to copy an existing OCLC microfilm record to a new record and edit out those elements which refer to the microfilm. The value of cloning can be illustrated by noting that the hit-rate in the early stages of the project was a little under 40% and that there were close copies for another 50%; in other words, without cloning, 60% of items would require substantial work, whereas with cloning this figure is less than 20%.

Simultaneously with the 'clone' command, ABN introduced an 'upgrade' command, which enables the direct conversion of an interim record into a full record; again, this can reduce the time required to catalogue an item by eliminating rekeying.

The Global Holdings Change facility, developed to deal with university amalgamations, is yet another recent development of benefit to the ABHR Project: it allows the addition of all holdings under a single sign-on without leaving them inaccessible to individual library catalogues. At the end of the project, all holdings listed under 'DBM:J' will be split off to the various individual institutions, which will then be able to download the records to their local systems.

Another recent development, the significance of which cannot be overstated, is the advent of AARNET.17 The benefits of this service to the ABHR Project are several. .

Firstly, it allows dial-up access to ABN without incurring the telecommunications charges associated with using Sim-PC.18 For ABHR staff based at Monash - accommodated in the Graduate Department of Librarianship, Archives and Records - this has been a major advantage: they are able to work through PCs with modems, instead of having to compete with the Department's students for the leased-line ABN terminals.

AARNET also allows access to a huge range of overseas databases. The ABHR Project has made considerable use of two major North American bibliographic utilities, OCLC and RLIN. 19 For purely technical reasons, Monash and the Fisher Library have used only RLIN; the National Library, which sampled both, found OCLC to be the more useful. The principal advantage of RLIN was that it allowed online access to the Eighteenth-Century Short Title


118 fan Morrison

Catalogue.20 As the project progressed the need to search overseas databases was reduced, as hit-rates on ABN improved (rising from 35-40% in early 1992 to 45-50% by mid-1993). The editing and authority work undertaken as an adjunct to the project also improved the general quality of the ABN database, increasing the proportion of original cataloguing that could be done without recourse to overseas databases. The gradual improvement in ABN hit-rates led to a deterioration in RLIN/OCLC hit-rates, as searches of these databases were restricted to increasingly obscure material: in the early stages of the project, RLIN hit rates were in the range 55-60%, but by late 1992 they had fallen to 45-50%.21

Cataloguing standards The articulation of acceptable minimum standards for records was a matter of some contention for EIP. 22 The ABHR Project has faced similar dilemmas, and a universally acceptable resolution of all the points at issue has proved impossible.

A significant difference between the standards adopted by the ABHR Project and those of EIP is the treatment of copy-specific data. ABN does allow for the recording of such information as 'Institution Specific Data'; but as the ABHR Project is entering data for all institutions under a single temporary customer number it cannot assign institution-specific data to the bibliographic records it creates, as such data would not be linked to any institution's customer number. Copy-specific information can be added to holdings statements - which will bp. transferred to individual institutions by a global holdings change - but the available space is extremely limited: the most that is possible is to note imperfections in the item, or other details of major significance which can be summarised within two lines (signatures or bookplates of previous owners, for example). . In general, if a record meets ABN standards it is acceptable to ABHR. The

difficulty is that there are a great many records on ABN, largely as a result of retrospective tape loads, that do not satisfy the letter or the spirit of ABN requirements. A great many of these have been upgraded as a result of the ABHR Project; a great many more have not. For example, University of New South Wales retrospective records (ABN rid Erefix 'nun') and South Australian EIP records (prefix 'eis'), both of which in different ways fail to satisfy the letter of ABN standards,23 have been edited only when new holdings have been added; it is probable that several thousand remain in need of editing.

The elements with which ABHR takes particular care are author entries, title transcriptions, imprints and collations. Lengthy titles and imrrints are truncated, but the aim is to give an accurate transcription of the text 0 the title-page. The first eight names in the imprint are recorded, with an indication of the number of omitted names. Titles are truncated on a more ad hoc basis: the only rule is to assess the significance of the information in the title and to effect no omissions that might limit the ability of a user to find or identify the item. Pagination is



reGorded scrupulously, with plates and other illustrative matter noted. Formats are preferred to measurements, but where the format is difficult to determine, measurements are acceptable. Care is taken to assign authorised forms of appropriate personal and corporate name headings.

ABHR's work in the vocabulary files falls into two categories: adding new headings and removing old or incorrect forms of headings. The first is a natural consequence of creating new bibliographic records. (In the case of ABHR, the vast majority of new headings to date have been those attached to ESTC items.) Removal of superseded or incorrect headings from the ABN vocabulary files occurs on a more ad hoc basis: authority work is part of ABHR's brief only insofar as it is necessary to create satisfactory bibliographic records. Nevertheless, the condition of the ABN vocabulary files, especially those pertaining to the Greek and Roman classics, is such that substantial authority work is a frequent occurrence. Classical authors 'tidied up' by ABHR include Tacitus, Plautus, and 'Longinus'. Swift, Defoe, Voltaire and Moliere are among the 'moderns' who have been worked on: in the case of Moliere, for example, 270 headings were reduced to 165, the spurious headings having been created by permutations and combinations of misspellings and misdatings, misapplication of diacritics, inappropriate MARC tags, failure to drop initial articles, use of English rather than French uniform titles, and the use of non-preferred forms of the author's name.

Subject access ABHR is not greatly concerned with subject access, or for that matter classification. Where appropriate Library of Congress subject headings exist on a record being cloned, they are taken over to the new record, but in general no effort is made to create new topical or geographic headings. Subject headings for persons, titles of works, and corporate bodies are generally added, as these are often uncovered in the course of authority work on author entries. There are, of course, cases where the title and imprint details are so nondescript as to demand some form of subject access: for example, the anonymous pamphlet' A letter to the Prime Minister. Dublin: [s.n.], 1748.' fairly begs for some further means of access beyond an ESTC reference. In this case the topical heading -'Ireland--Politics and government--18th century' - was assigned.24

It is arguable that a work has not been fully described if its intellectual content has not been analysed. It is also arguable that, flawed as they are, Library of Congress Subject Headings are the best available source of descriptions of intellectual content for library cataloguing purposes. The ABHR Project, however, has worked on the assumption that subject entries are not the usual means of access to early printed books: most commonly the requirement will be for a specific title, or an author or printer or publisher.15

A library catalogue, in whatever form, fulfils a function different from that of specialised bibliographies, enumerative or descriptive. A library catalogue is


120 fan Morrison

an aid in the construction of specialised bibliographies and does perhaps have a certain amount in common with enumerative bibliography, but a catalogue is first and foremost a finding aid. It was decided, therefore, that subject headings were dispensable. A great many of the existing records for pre-1801 material on ABN lack subject headings, and the time required for full subject cataloguing would mean abandoning all hope of anything like complete recording. At the time of writing the ABHR Project could, with the compromises it has made (both with EIP standards for copy-specific data and with subject access), realistically aim to achieve a 90% coverage of pre-1801 items in Australian libraries: to have rejected those compromises would have reduced the likely coverage to something under 50%.

Outcomes It is important to recognise what the ABHR Project is not. ABHR will not fulfil bibliographers' dreams of a complete, internally consistent, and copiously detailed listing of every piece of printing carried out before 1801 now held in Australia. Significant compromises have been made in the quality of the records input in order to achieve the widest possible coverage. (It should also be stressed that even attaining these minimal standards led to a substantial improvement of the ABN database.) AB HR is a major advance in an ongoing process, but much work will remain to be done. The projected CD-ROM, for example, will require a great deal more than a simple extraction of records from ABN. Considerable editing work may be required to achieve uniform standards in the records and to eliminate duplication; a CD-ROM might be produced without this work being done, but the less preliminary editing that is done the less predictable - and therefore less malleable - the data will be.

Nonetheless, the ABHR Project will have added tens of thousands of medium to high quality bibliographic records to ABN. It will have added holdings statements to tens of thousands more existing bibliographic records. It will have enhanced, upgraded or amended thousands more bibliographic records. In doing all this, it will also have effected a substantial improvement in the ABN vocabulary files, in particular the name authorities for author entries. (A more distant effect will be the enhancement of local systems, as the institutions whose holdings were recorded download records.)

ABHR will make the job of rare books cataloguers easier. There is considerable scope for further enhancement of the records created by AB HR and those left unedited by ABHR, and there are grounds for optimism that this will occur as cataloguers find less of their time being consumed by the demands of original cataloguing.

An outcome of wider significance will be to make collections of early printed material in Australian libraries more generally known. Although the ABHR Project will fall short of 100% coverage, it will be able to give an accurate account of the material not listed on ABN. A detailed account of collections still left



unrecorded will be prepared at the completion of the project, but the approximately 10% of Australian holdings not listed on ABN at the completion of the ABHR Project will largely comprise the following:

private collections;26 collections at institutions that have not responded to approaches from ABHR for permission to add their EIP records;27 very small collections in outlying areas;28 EIP records from Western Australia, Tasmania and Queensland;29 a total of several thousand items at the Fisher Library, University of Sydney, and the State Library of New South Wales; items too fragmentary to be properly identified or usefully catalogued;30 serials and engraved material (e.g. maps and music);31 a collection of appeal cases ~approximately 1000 separate items) at the Supreme Court of Victoria.3

In short, with a few readily definable exceptions, all the pre-1801 material in New South Wales, the Australian Capital Territory, Victoria, and South Australia will be recorded on ABN. Researchers will be able to ascertain the presence, and locate copies, of about 90% of the early printed books held in Australian libraries, with one call to the library at their home institution.33 By using Supersearch, it will also be possible to extract records of works produced in a given place, or by a certain printer, or in a particular year, or combinations of these. Local systems that download records from ABN will also be greatly enhanced. As AARNET access to local systems expands (a great many are already accessible) avenues for research can only be enlarged.

The ABHR Project is a major advance in the ongoing process of enhancing access to early printed material in Australian libraries.

Melbourne

NOTES

1. AIan Brissenden, 'EIP in South Australia', Australian Library Journal 30,2{1981}: pp.43-6; B.J.McMullin, 'The Australia and New Zealand Early Imprints Project: the background', BSANZ Bulletin 6,4{1982}: pp. 163-73.

2. A blank EIP card is reproduced by McMullin, pp.170-71. 3. McMullin, pp.167-68, lists the executive committee and the chairmen of the various regional

committees at the height of EIP activity in 1982. 4. A. W. Pollard & G .R. Redgrave, A short-title catalogue of books printed in England, Scotland and

Ireland and of English books printed abroad 1475-1640, 2nd ed., rev. by W.A. Jackson, F.S. Ferguson, & K.F. Pantzer {London: Bibliographical Society, 1976-86}.

5. D. G. Wing, Short-title catalogue of books printed in England, Scotland, Ireland, Wales, and British North A merica and of English books printed in other countries 1641-1700, 2nd ed. (New York: Modern Language Association of America, 1972-88).

6. MARC: Machine Readable Cataloguing. A standard format for recording catalogue data in electronic form, developed during the 1960s and 1970s. There are many regional variations in


122 fan Morrison

MARC formats, although all are broadly compatible. A record in the 'AUSMARC' format used by ABN is shown on p.124.

7. John Fletcher and Rose T. Smith, A short· title catalogue of sixteenth century printed books held in libraries and private collections in New South Wales (Sydney: Library Council of NSW, 1979). Brissenden, p.44, notes the 'considerable advantage' given New South Wales by Fletcher and Smith.

8. Mechanism C funds infrastructure projects, as distinct from pure research. 9. Except to the extent that elements from them (typically publication dates) were entered in

uniform titles and the fixed fields of the MARC record. In recent years the development of Supersearch, which uses more powerful retrieval software, has overcome many of the limitations of the WLN-based system.

10. R.K. Olding, 'Technical services' , A n enthusiasm for libraries: essays in honour of Harrison Bryan, ed. Jean P.Whyte and Neil Radford (Melbourne: Ancora Press, 1988), p.12S.

1l. ABN observes the international standards set out in the Anglo·American Cataloguing Rules, 2nd ed., 1988 revision (Chicago: American Library Association, 1988).

12. CA V AL (Co-operative Action by Victorian Academic Libraries) provides cataloguing services on a contract basis.

13. South Australian EIP records on ABN can be recognised by rid (record identifier) numbers with the prefix 'eis'.

14. Online Computer Library Center, one of the major North American bibliographic utilities. 15. Storage and Information Retrieval Systems: an information retrieval system developed by IBM,

familiar to most librarians with experience of on-line searching. Note: ST A1R.S can be used only for retrieval; the WLN software is still required for data entry.

16. With no guarantee of success: until the advent of Supersearch there was little incentive to seek out bibliographic references, and even less to standardise forms. (If 'STC' in the required number has been entered as'S. T. C.', a search specifying 'STC' would fail.)

17. The Australian Academic and Research Network, established and funded on a co-operative basis by Australian universities; now increasingly being made available on a cost-recovery basis to other institutions.

18. A software package used for access to ABN via AUSTPAC. 19. The Research Libraries Information Network, the bibliographic network of the Research

Libraries Group, based in Stanford, California. The RUN database includes a number of separate special databases; the one of particular interest to ABHR is the Eighteenth Century Short Title Catalogue.

20. ABHR has had some communication, but no formal collaboration, with ESTC. 2l. The ABHR experience of RUN and OCLC cannot be generalised too freely, as these databases

were searched only for items for which no record, or close-copy record, was found on ABN. The selection of items to search on RLIN / OCLC left considerable scope for variation: although ABHR procedures made original cataloguing a third step after searching ABN and RLIN/ OCLC, the RLIN/OCLC search might well be bypassed where a verified heading existed on ABN and the item presented no bibliographic difficulties, even though no close-copy record existed on ABN. (The commonest items in this category were Wesleyan sermons, Greek and Roman classics, and English plays.) The National Library used OCLC chiefly for the residue of uncatalogued material left after its retrospective conversion, obscure even by the standards of rare books cataloguing, and hit-rates varied considerably from collection to collection, as follows: Alston and Kashnor 42%; Valente 20.5%; French Revolutionary Pamphlets 16.4%. (The Valente and French Revolution collections had hit rates of 2.2% and l.3% respectively on ABN.) The greatest discrepancy between ABN and RLIN/OCLC was in the National Library's Dutch Pamphlets: 0% on ABN, 57.9% on RUN.



Another variable is the effect of the project in enhancing the skills of the staff employed on it. As competence and confidence levels increased, psychological dependence on overseas databases was reduced.

22. See McMullin, p.169, for a summary of the discussion of appropriate standards. 23. University of New South Wales records are usually very brief, with truncated imprints and

often no statement of pagination or dimensions -let alone format - and non-AACR2 forms for headings. South Australian EIP records often use non-AACR2 headings, contain detailed copy-specific information (for example, descriptions of bindings), and other technical flaws resulting from minor discrepancies between the version of MARC used for the original EIP records and that used by ABN.

24. Another entry might be the name of the Prime Minister addressed, but the point is to minimise the time spent on each individual record. This is one case where an appropriate topical heading was easier to identify than a personal name.

25. Many Library of Congress Subject Headings - other than for history and literature - are unsatisfactory as a means of describing the content of early printed books. The subdivision 'Early works to 1800', for example, is less than helpful when applied - as it often is - to subjects such as astrology that had expired as respectable scholarly disciplines by the early nineteenth century.

26. Although the EIP SA tapes include material in private hands 'sponsored' by the State Library of South Australia.

27. Most of these are very small collections: they account in total for less than 3% of ElF records held at Monash.

28. Where the cost of recording (including the opportunity cost in terms of recording larger collections) outweighs the benefit.

29. Due to difficulties in converting these records into MARC format. About 10,000 records are involved.

30. These represent between 3 and 4% of EIP records. 31. These were outside the scope of both EIP and ABHR. 32. Recorded by EIP but put aside by ABHR on purely pragmatic grounds: the amount of

authority work required to adequately catalogue these items (they are all privately printed) seems out of all proportion to the use likely to be made of them.

33. Access to ABN, however, is available only through its member libraries: researchers who are not practising librarians will not readily be able to conduct their own searches. The production of a CD-ROM has the potential to overcome this limitation, but the twin requirements of familiarity with software and access to hardware would remain.

BSANZ Bulletin, v.17 00.3, 1993

124 fan Morrison

ABN RECORD IN MARC FORMAT

Fields marked * are searchable with WLN software; but note that the Cc' subfield in the TfL (title) field, which transcribes the statement of authorship on the title-page, cannot be searched.

* * *

* *

am abn93-167817 db 05/18/93 --/--/- 05/19/93 DBM:J DBM:J MEPF I acd I Moses, I of Khoren, 15th cent. UTIAO I al I Patmut'iwn Hayots'.1 Latin & Armenian TILAO I abc I Patmut'iwn Hayoc' ew Asxarhagrutiwn. : I Mosis

Chorenensis Historire Armeniacre libri III ... / I Armeniace ediderunt, Latine verterunt, notlsque illustrarunt Gulielmus & Georgius, Gul. Whistoni ...

IMP I abc I Londini : I Ex officina Caroli Ackers typographi; apud J oannem Whistonum bibliopolam, 11736.

COL abc {4}, xxiii, {1}, 412 p., plate: I map ; I (4to) NOG a Parallel Latin and Armenian text. NOIL ac ESTC,It147675 SUG-L axy Armenia I History ITo 428. AEPSA ad Whiston, William, I 1667-1752. LAN a latarm CAS ac DBM:JIDBM:J LON a (atABN) 10028656 IMM a abn93165281

FFD CONF= ME IN B= x INTEL LV= DAT KY = s LAN= lat CNTRY =enk ILLUS= b MODRC= CATS= d

FEST= INDEX = FIC= BIOG= DATE1= 1736 DATE2= REPRO= CON TEN T S = GOVPUB= CAT FORM = a


Copyright of Full Text rests with the original copyright owner and, except as permitted under the Copyright Act 1968, copying this copyright material is prohibited without the permission of the owner or its exclusive licensee or agent or by way of a license from Copyright Agency Limited. For information about such licences contact Copyright Agency Limited on (02) 93947600 (ph) or (02) 93947601 (fax)

Documents

THE AUSTRALIA'S BOOK HERITAGE RESOURCES PROJECT€¦ · · 2015-05-12THE AUSTRALIA'S BOOK HERITAGE RESOURCES PROJECT IAN MORRISON ... but only South Australia used a standard MARC